Rome talk Carolina

Slide1: 

Exploring the potential of γ/Z+jet events for In-Situ calibration A.Gupta, M.Hurwitz, S.Jorgensen, B.Kehoe, J.Proudfoot, C.Deluca JetETMiss ATLAS Physics Workshop, June 6-13, 2005

Slide2: 

Outline Event samples Global /Z + Jet events event properties Methods to study jet energy scale Relative jet energy scale across detector Absolute energy scale Particle level studies jet energy scale (EM and H1-weighting) Comparing Z and  Dijet background to photons Conclusions

Slide3: 

Introduction Motivations  or Z0 is a well calibrated object at EM scale balancing the recoiling hadronic system potentially large statistics available pT range from 20 GeV to ~60 GeV: Z(ll)+jet (~2Hz) γ+jet (~ 0.1 Hz) reserving 1Hz for downscaled trigger pT range > 60 GeV: γ+jet (~2Hz) Z+jet (~ 0.1 Hz) Issues to be understood Detector effects: response, showering Physics effects: fragmentation, gluon radiation (multijets) Want to compare different analyses in these events

Slide4: 

Event Samples Z+jet Zee and Z - Rome sample with Athena 10.0.1 but also DC2 / 9.0.3 and DC2 / 10.0.1 γ+jet pT > 20GeV signal & dijet background - DC1 sample with Athena 7.2.0 pT > 60 GeV signal & dijet background - Rome sample with Athena 10.0.1

Slide5: 

Zmass Zee:M=91.79GeV Zmm:M=91.49GeV Rome data vs CDF data Global event properties Jet Multiplicity · in Z 40% higher than in Zee · higher in 9.0.3 (DC2) - more forward jets #jets

Slide6: 

Pt balance Calculated from leading jet and photon/Z Sensitive to out-of-cone showering, underlying event, gluon radiation We want to study, and factor out from detector response Etmiss Projection Etmiss (vector sum over everything in calorimeter) Sensitive to particle response only, not effects from algorithm applied to recoil Etmiss in following plots calculated at tower level Methods to Study Jet Energy Scale

Slide7: 

Recoil made largely of jets… Leading Jet Z,jet distribution Response (R) and Z Event Topology Z,jet Z,jet Dependence of R on Z,jet - relatively insensitive if jet in opposite hemisphere No Dependence of R on #jets # jets

Slide8: 

Jet Response Vs. E’ = ETZ*cosh(jet) Pt balance (blue lineEM scale) Both are ~20% lower in the EM scale Absolute scale of Leading Jet at EM scale E’ leading jet R

Slide9: 

pTbalance jet (E) 10.0.1 Rome Jet response (Etmiss projection) Zee Relative calibration across the detector at EM scale see dips at eta 1.5 and 3.2, and perhaps 0.8 Expected from particle-level comparisons (F. Paige/S. Padhi) Pt balance flatter in eta E (and Response) is increasing with η: both R and balance see this out-of-cone showering also increases with eta: only balance sees that effect  cancels higher response giving flatter balance (E;  ) jet

Slide10: 

pTbalance calibrated jets |h|<2.5 |h|>2.5 Zee Relative calibration across the detector at HAD scale Calibration or Reconstruction problem for |h| > 2.5 ? More study needed.

Slide11: 

Absolute energy scale - particle level studies (1)  +jet events - jets reconstructed at particle level Low ET: <jetET>~30 GeV

Slide12: 

Absolute energy scale - particle level studies (2) Higher ET: <jetET>~70 GeV Issues Understand different jet algorithms Energy dependence

Slide13: 

Calibrated rec. jet energy / Particle level jet energy Hadronic jet energy calibration (H1-weighting) Issues: few % difference in calibration for different jet algoritms MEAN

Slide14: 

Zee Zee:Balance = -0.109 Z: Balance = -0.111 Compared with  Offset of approx. 0.05 relative to Z+jet Comparing Z and  Statistical error ~ 0.3%

Slide15: 

Dijet background to  sample Low pT High pT Statistical error Remaining jet background  π0’s

Slide16: 

Conclusions Real data and MonteCarlo simulation will be systematically compared once real data-taking starts This will allow to cross-check our understanding of a lot of physics (QCD , fragmentation, etc.) effects our correct detector description in the Monte Carlo relative jet algorithm behaviour If Data and MC compare well, then one can start thinking of using the process for precise absolute energy calibration